CN104010340A - Urban vehicle network message multicast routing method based on node movement trend - Google Patents

Abstract

Provided is an urban vehicle network message multicast routing method based on a node movement trend. A target area dividing method based on the movement trend is provided according to a model built for urban road traffic and a communication network, and according to the movement state of nodes, the nodes are grouped dynamically in real time, and routing activities are arranged. Due to the fact that the actual road condition is considered, a multicast routing algorithm is more suitable for diffusion of organization messages in an urban vehicle network, the extra cost of the network can be greatly lowered on the premise of not obviously reducing the delivery success rate, and most of (80 percent) of the messages can be delivered to the position of a target node within a short time (about 300 seconds).

Description

A kind of city vehicle internet message multi-broadcast routing method based on node movement tendency

Technical field

The present invention relates to network communication field, particularly the message in vehicle network forwards and diffusion problem, is a kind of message multi-broadcast routing method specifically, is applicable to the Information Communication of city vehicle network.

Background technology

Vehicle network (Vehicular Networks) refers to that the electronic devices and components that are loaded on vehicle are by recognition technologies such as less radio-frequencies, realization on information network platform to the attribute information of all vehicles and quiet, multidate information extracts and effectively utilize, and according to different functional requirements, the running status of all vehicles is effectively supervised and provided a class network of integrated service.Important application as technology of Internet of things in intelligent transportation system (Intelligent Transportation System, ITS), vehicle network technology has higher technology and economic feasibility, has very wide prospect.

The network that intelligent transportation system relies on and the network that traditional node is fixed, link change is less have very large different, particularly when during the situations such as between complicated road conditions, vehicle the dynamic change of link topological structure is violent, how ensureing that the communication between each sensing, communication node becomes one of key issue building City ITS in the face of city.Researcher is by introducing delay-tolerant network (Delay Tolerant Networks, DTNs) technology solves this problem, the transmission that the DTNs utilization mechanism of " storing-carry-forward " completes message is (referring to Chuah M C, Ma W.Integrated buffer and route management in a DTN with message ferry[C] //Military Communications Conference, 2006.MILCOM2006.IEEE.Washington, DC, IEEE, 2006:1-7).The essence of this mechanism is by introducing time dimension, no longer only utilize the space connected relation of bottom link topology to carry out the communication of data, but the connected relation of comprehensive utilization bottom link in two dimensions of room and time carries out data transmission, thereby can utilize more communication opportunity, by the low connectedness on single Spatial Dimension, intermittent communicated property structure, be extended for the strong connectedness in two dimensions of space time, continue connective structure, strengthened network communications capability, make under vehicle network environment, network still can provide reliable delivery of packets service for upper-layer protocol.

Meanwhile, along with mobile self-grouping network (Mobile Ad-hoc Networks, MANETs) and vehicle self-organizing network (Vehicular Ad-hoc Networks, VANETs) development, application in vehicle network is increasingly extensive, the message content exchanging between vehicle node also becomes increasingly abundant, not only be confined to the panic alarm information based on traffic system such as road conditions, accident, even comprise various application entertainment information, commercial advertisement etc. in position-based service system (Location Based Service, LBS).Be not difficult to find out, due to the particularity of vehicle network, these information all with the position of vehicle, move closely related.For example, user is concerned about the traffic congestion information in Vehicle Driving Cycle the place ahead more, or near the relevant information such as parking lot, gas station, restaurant vehicle current location.And in research in existing vehicle network, lacking the consideration to the different information preference of user, the Routing Protocol in traditional vehicle network can not effectively be distinguished according to the content of message, thereby causes the waste of mass communication resource.

Routing Protocol in existing vehicle network mainly contains two kinds of mode classifications: according to route selecting information and according to route transmission strategy (referring to Li F, Wang Y.Routing in vehicular ad hoc networks:A survey[J] .Vehicular Technology Magazine, IEEE, 2007,2 (2): 12-22.).According to route selecting information, the Routing Protocol in car networking can be divided into two large classes, be respectively based on topology information and position-based information.In agreement based on topological, node is more responsive to network topology, and forwarding messages mainly utilizes the node in connection status in network; In location-based agreement, more responsive to the positional information of node.Route transmission strategy is delivered to destination node for guide message from source node, according to the quantity of destination node and have or not grouping situation can be divided into following four classes: without (the clean culture of grouping simple target node, Unicast), (choosing is broadcast to have the interior simple target node of grouping group, Anycast), without (the broadcast of grouping multiple target node, Boardcast), have multiple target node in grouping group (multicast, Multicast).Multicast protocol again can according to the information difference of grouping institute foundation be divided into the auxiliary multicast (GeoCast) in geographical position (referring to Maihofer C.A survey of geocast routing protocols[J] .Communications Surveys and Tutorials, IEEE, 2004, 6 (2): 32-42.), social property assists multicast (SocialCast) (referring to Deng X, Chang L, Tao J, et al.Social profile-based multicast routing scheme for delay-tolerant networks[C] //Communications (ICC), 2013IEEE International Conference on.IEEE, 2013:1857-1861.) etc.

The grouping of multicast protocol is according to being all the attribute of communication node after all, comprise build-in attribute (as, social property) and time become attribute (as, geographical position attribute) etc.Wherein especially with node time become attribute and be most widely used in the network of this node high-speed mobile of vehicle network, while as motion states such as the position of node, speed etc. being all, become.These information relevant with geography information or node motion state play vital effect to the accurate delivery of message.In existing research, with geographical location information, receive publicity the most.In addition, while also having based on node empty, information is carried out multicast, and the agreement of sign is exactly mobile real-time multicast (MobiCast).Utilize the trend characteristic group technology of node also to have research to relate to, the people such as Allal have proposed target area (ZOR) can be divided into some child partitions (sub-ZORs), and the division of child partition has common trend objective (referring to Allal S according to being them, Boudjit S.Geocast Routing Protocols for VANETs:Survey and Guidelines[C] //Innovative Mobile and Internet Services in Ubiquitous Computing (IMIS), 2012Sixth International Conference on.Palermo, 2012:323-328), the people such as Chen also utilize the trend characteristic of node, the node of judgement trend " incident area " is destination node, carry out multicast transmission accident information (referring to Chen Y, Lin Y, Lee S.A mobicast routing protocol in vehicular ad-hoc networks[J] .Mobile Networks and Applications, 2010,15 (1): 20-35.).Yet in these schemes, the target area that the decision-making of movement tendency institute goes out is static, when the realizing of agreement, need to be serviced, can not effectively be applied in large-scale city vehicle network environment.

Summary of the invention

Technology of the present invention is dealt with problems: for the deficiency of multicast route protocol in existing vehicle network, propose a kind of based on the auxiliary message multi-broadcast routing method of node movement tendency, by introducing the DTNs communication technology, adopt the proper communication in frequent interruption situation between vehicle in " store-carry-forward " machine-processed support vehicles network; Utilize the movement tendency of vehicle node dynamically to judge whether the target area in message, and then organize targetedly routing activity, avoided the huge waste of the communication resource.

Technical solution of the present invention: a kind of city vehicle internet message multi-broadcast routing method based on node movement tendency, specific implementation step is as follows:

(1) whether vehicle node cycle criterion self sets up state or direction of motion variable condition in link, if set up state in link, and execution step (2), if in direction of motion variable condition, execution step (3);

(2) vehicle node and some nodes are set up communication linkage, carry out message sub-process, and concrete grammar is:

(2.1) resource that initial message transmission needs, resource comprises obtains the queue of message sending and receiving, and obtains the essential information of state of motion of vehicle;

(2.2) all-links that traversal vehicle node is set up, to each link execution step (2.3), until traversal finishes;

(2.3) transmit queue of recipient's node at traversal link two ends, to each message packets execution step (2.4), until traversal finishes;

(2.4) which kind of region that the vehicle node present position of judgement message receiver is message, processes according to judged result, is specially:

If (2.4.1) target area in message, receives this message, put into ACCEPT queue (message sink queue);

If (2.4.2) Relay Region in message, receives this message, put into CARRY queue (message is carried queue);

If (2.4.3) extraneous areas in message, ignores this message.

(3) vehicle node seesaws direction while changing, XM cache management sub-process, and concrete grammar is:

(3.1) resource that initialization cache management needs, resource comprises obtains message queue, and obtains the essential information of state of motion of vehicle, the change type of judgement direction of vehicle movement;

(3.2) buffer queue of traversal vehicle node, to each message execution step (3.3), until traversal finishes;

(3.3) if the change type of direction of vehicle movement is oppositely tune, carry out (3.3.1); If the change type of direction of vehicle movement is left and right, turn, carry out (3.3.2); If the change type of direction of vehicle movement still, for keeping straight on, is left intact;

(3.3.1) the residing ACCEPT queue of this message or CARRY queue are exchanged;

(3.3.2) reduce the quota of this message covering path quantity, the message digit that replacement movement tendency calculates is set to the position when vehicle in front, then judges the current region of living in of vehicle node is for which kind of region of this message: if ACCEPT queue is inserted in target area; If Relay Region, inserts CARRY queue; If extraneous areas, deletes this message.

Wherein, which kind of region that in step (2.4), judgement vehicle node present position is message, concrete grammar is:

1) determine that vehicle node position is to the orientation vector v of message source node position ₁, determine vehicle node direction vector v now ₂, obtain the angle theta between two vectors;

2) according to the directivity feature between two vectors of size judgement of angle theta, be specially:

2.1) if θ is ∈ [0, π/6], two vectors are considered as in the same way; If θ is ∈ (π/6, π/3), execution step (2.2); If θ is ∈ [π/3,2 π/3], two vectors are considered as intersecting; If θ is ∈ (2 π/3,5 π/6), execution step (2.3); If θ is ∈ [5 π/6, π], two vectors are considered as oppositely;

2.2) utilize angle between two vectors of Probit model decision-making in the same way or intersect;

2.3) utilizing the angle between two vectors of Probit model decision-making is intersection or reverse;

3) according to two vector v ₁and v ₂between directivity feature, obtain the taxis feature tendency of vehicle node and message, be specially:

3.1) if v ₁with v ₂in the same way, vehicle node trends towards message source node position;

3.2) if v ₁with v ₂oppositely, vehicle node deviates from message source node position;

3.3) if v ₁with v ₂intersect, vehicle node has nothing to do in message source node position;

4) obtain the set TENDENCY of the taxis preference of message _m(being the attribute of message, along with the product of message is born with), judge whether the taxis tendency of vehicle node and message source node position belongs to this set, thereby which kind of region that the vehicle node present position of judgement message receiver is message is specially:

4.1) if tendency ∈ is TENDENCY _m, the target area in message;

4.2) if～tendency ∈ TENDENCY _m, the Relay Region in message;

4.3) otherwise, the extraneous areas in message.

Meanwhile, the change type of judgement direction of vehicle movement in step (3.1), concrete grammar is:

1) determine the direction of motion vector v of vehicle node previous moment ₁, determine vehicle node direction vector v now ₂, obtain the angle theta between two vectors.

2) according to the change type of the size judgement direction of vehicle movement of angle theta, be specially:

2.1), if θ is ∈ [0, π/6], by the change type of looking direction of vehicle movement, be craspedodrome; If θ is ∈ (π/6, π/3), execution step (2.2); If θ is ∈ [π/3,2 π/3], by the change type depending on direction of vehicle movement, be to turn in left and right; If θ is ∈ (2 π/3,5 π/6), execution step (2.3); If θ is ∈ [5 π/6, π], by the change type of looking direction of vehicle movement, be oppositely tune.

2.2) utilize the change type of Probit model decision-making direction of vehicle movement for keeping straight on or left and right turning.

2.3) utilize the change type of Probit model decision-making direction of vehicle movement for turning or oppositely tune left and right.

In addition, repeatedly utilized Probit model to carry out decision-making in the present invention, Probit model can be described to:

$P_i=\mathrm{Prob}(Y_i=1|x)=\mathrm{Prob}[u_i>-({\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}})]=1-F[{\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}}]$

Wherein, X _ijj independent variable while representing the i time prediction, Y _idependent variable while representing the i time prediction, P _ithe probability that during the i time prediction, dependent variable value is 1, β ₀, β _j, μ _ifor model parameter.In the present invention, the explanatory variable of choosing is angle theta (ANGLE), can suppose that regression model is:

Y _i＝F ^-1(P _i)＝β ₀+β ₁ANGLE _i+μ _i

Wherein, during be message in judgement vehicle node present position which kind of region, have: and when the change type of judgement direction of vehicle movement, have:

Finally, utilize the method for regression analysis, by some actual acquisition to data calculate corresponding parameter, then carry out the corresponding result of decision-making with this model.

The present invention's advantage is compared with prior art: combine the actual conditions of urban road, the target area of message is set more neatly, dynamically utilize the movement tendency assistance messages diffusion of vehicle node, save mass communication resource.Specifically, first, consider transmission of news characteristic in vehicle network, according to the expectation propagation path of message, message is classified, different classes of message is adopted to different route replies, and assurance message is delivered more quickly and accurately.Secondly, utilize vehicle node dynamically to judge whether the target area in message with respect to the movement tendency of message source position, and then carry out routing forwarding targetedly, saving resource, has also reduced the maintenance costs to target area simultaneously.

Accompanying drawing explanation

Fig. 1 is the flow chart of message Multicast Routing Algorithm in the present invention;

Fig. 2 is the principle schematic that judges two directivity features between vector in the present invention;

Fig. 3 is that the present invention is applicable to the principle schematic that road traffic accident early warning information is propagated;

Fig. 4 is that the present invention is applicable to the principle schematic that message propagation is dodged in emergency vehicle request;

Fig. 5 is the message dilivery success rate performance comparison diagram of the present invention and traditional multicast route;

Fig. 6 is the message overhead rate performance comparison diagram of the present invention and traditional multicast route;

Fig. 7 is the message delay distribution performance comparison diagram of the present invention and traditional multicast route.

Embodiment

City vehicle network is comprised of urban highway traffic participation device, and the urban road of take is carrier is carrier serves the current a kind of latticed form of urban road.These participation devices comprise: carry vehicle, the pedestrian of communication terminal, and the infrastructure that roadside is fixing etc., but each equipment has, message produces and the function of route.

One, introduce below basic embodiment of the present invention.

1, first, need to carry out modeling to network environment.The content of modeling comprises: the target area of road, vehicle node, message, message, the Relay Region of message etc.Specific as follows:

(1) urban road and related content modeling

In the present invention, the concept of definition position and location sets is as follows:

$\left\{\begin{array}{c}\mathrm{location}=(\mathrm{longitude},\mathrm{latiude})\\ \mathrm{LOCATION}::=\left\{p\right|p=\mathrm{location}\}\end{array}\right.$

Wherein, location represents position, and LOCATION represents location sets, the locative longitude of longitude, the locative latitude of latitude.

The concept that defines road and road set in the present invention is as follows:

$\left\{\begin{array}{c}\mathrm{road}=({\mathrm{id}}_r,\mathrm{locatio}{n}_{\mathrm{begin}},{\mathrm{location}}_{\mathrm{end}},{\mathrm{direction}}_r,\mathrm{width})\\ {\mathrm{location}}_{\mathrm{begin}}\∈\mathrm{LOCATION}\\ {\mathrm{location}}_{\mathrm{end}}\∈\mathrm{LOCATION}\\ {\mathrm{direction}}_r\∈\{\mathrm{Antero}\mathrm{grad}e,\mathrm{Retro}\mathrm{grad}e,\mathrm{Bidirectional}\}\\ \mathrm{ROAD}::=\left\{r\right|r=\mathrm{road}\}\end{array}\right.$

Wherein, road represents road, and ROAD represents the set of road, id _rthe uniqueness sign that represents road, location _begin, location _endthe terminal that represents road, direction _rthe direction that represents road, width represents the width of road.The terminal of road is all a location point, and road direction is divided into three kinds: forward (Anterograde), reverse (Retrograde) and two-way (Bidirectional).The look out for the car same tropism of flow path direction and origin and destination direction of the standard of judgement direction, if wagon flow is from road origin-to-destination, for forward; Be reverse on the contrary; If two-way unseparated road is two-way.

The concept that defines geocoded activity in the present invention is as follows:

Wherein, geocoded activity be used for judging a location point be (True) no (False) in certain section, be used for judging any the bar road of a location point in road set.

In the present invention, reverse road activity is asked in definition concept as follows:

Wherein, road _anterograderepresent forward road, road _retrograderepresent reverse road, road _{bidirectional}represent two-way road, asking for reverse road object is that acquisition is arranged side by side with current road, but the road of wagon flow opposite direction.

(2) vehicle node and related content modeling

The concept that defines vehicle and vehicle set in the present invention is as follows:

$\left\{\begin{array}{c}\mathrm{vehicle}=({\mathrm{id}}_v,{\mathrm{location}}_v,{\mathrm{direction}}_v,\mathrm{speed},\mathrm{comm}\_\mathrm{info})\\ {\mathrm{location}}_v\∈\mathrm{LOCATION}\\ {\mathrm{direction}}_v\∈\{\mathrm{Forward},\mathrm{Backward},\mathrm{Stationary}\}\\ \mathrm{comm}\_\mathrm{info}=(\mathrm{radius},\mathrm{bandwidth},\mathrm{buffer})\\ \mathrm{VEHICLE}::=\left\{v\right|v=\mathrm{vehicle}\}\end{array}\right.$

Wherein, vehicle represents vehicle, and VEHICLE represents the set of vehicle, id _vthe uniqueness sign that represents vehicle, location _vthe current location that represents vehicle, direction _vrepresent the current travel direction of vehicle, speed represents the travel speed that vehicle is current, and comm_info represents the communication information of vehicle, and radius represents the communication radius of vehicle, and bandwidth represents the communication bandwidth of vehicle, and buffer represents the communication buffer of vehicle.Vehicle node mainly comprises two aspect information, the mobile message of vehicle and the communication information of node, and wherein current location and traffic direction are two main mobile messages; The communication information comprises communication radius, bandwidth, cache size etc.Traffic direction, mainly for wagon flow direction, is divided into direct motion (Forward), retrograde (Backward) and static (Stationary) three kinds.

In the present invention, movement tendency judgement is an activity of outbalance, and position and traffic direction according to when vehicle in front, judge that it is with respect to the taxis feature of fixed position.The judgement of definition trend concept as follows:

Wherein, TENDENCY represents the set of trend state.As shown in Figure 2: if θ is ∈ [0, π/6], two vectors are considered as in the same way; If θ is ∈ (π/6, π/3), utilize angle between two vectors of Probit model decision-making in the same way or intersect; If θ is ∈ [π/3,2 π/3], two vectors are considered as intersecting; If θ is ∈ (2 π/3,5 π/6), utilizing the angle between two vectors of Probit model decision-making is intersection or reverse; If θ is ∈ [5 π/6, π], two vectors are considered as oppositely.If v ₁with v ₂in the same way, vehicle node trends towards (Tended) message source node position; If v ₁with v ₂oppositely, vehicle node deviates from (Deflected) message source node position; If v ₁with v ₂intersect, vehicle node has nothing to do in (Irrelevant) message source node position.

(3) message and related content modeling

The concept that defines message and massage set in the present invention is as follows:

$\left\{\begin{array}{c}\mathrm{message}=({\mathrm{id}}_m,{\mathrm{location}}_m,{\mathrm{TENDENCY}}_m,\mathrm{source},\mathrm{range},)\\ \mathrm{ttl},\mathrm{type},\mathrm{quot}{a}_{\mathrm{change}}\\ {\mathrm{location}}_m\∈\mathrm{LOCATION}\\ \mathrm{source}\∈\mathrm{LOCATION}\\ {\mathrm{TENDENCY}}_m\∈\mathrm{TENDENCY}*\\ \mathrm{content}=(\mathrm{topic},\mathrm{description})\\ \mathrm{MESSAGE}::=\left\{m\right|m=\mathrm{message}\}\end{array}\right.$

Wherein, message represents message, and MESSAGE represents the set of message, id _mthe uniqueness sign that represents message, location _mrepresent the position of message for calculating movement tendency, TENDENCY _mrepresent the set of message taxis preference, source represents the source position of message, and range represents the coverage of message, and ttl represents the life span of message, and type represents the type of message, quota _changerepresent the limit that message can covering path quantity, priority represents the priority of message, size represents the size of message, content represents the content of message, comprise message theme (topic), (description) etc. described, content partial content is the message content that upper layer application is paid, and other information are the information that routing decision need to be used.The reference position of the taxis judgement of the relative message of vehicle is location _m, but not source, location _mbe initially source, after node turns to, understand mark location again _mfor the position of turning point, such processing contributes to the judgement of taxis more accurate.

Message expectation receiving node set E in the present invention _mconcept definition as follows:

Wherein, message expectation receiving node is the destination node of not considering the message in road constraint situation.

(4) message relevant range modeling

In the present invention, define targets of messages region TR _mwith target vehicle TV _mconcept as follows:

Wherein, targets of messages zoning, the vehicle in this region, section is all the target vehicle of message.

In the present invention, define message Relay Region FR _mwith forwarding vehicle FV _mconcept as follows:

Wherein, message Relay Region stipulated, the vehicle in this region, section is all the forwarding vehicle of message.

2, secondly, determine the multicast routing policy of message and the strategy of cache management.This part is core of the present invention, as shown in Figure 1, in the circulation of carrying out, need to process two sub-processes in normal agreement, is respectively message sub-process and cache management sub-process.Specific as follows:

(1) message sub-process

Message sub-process in the present invention may be defined as:

Wherein, STRATEGY represents the set of routing activity, comprises reception (Accept), carries (Carry) and ignores (Ignore).Concrete steps are:

(1.1) resource that initial message transmission needs, resource comprises obtains the queue of message sending and receiving, and obtains the essential information of state of motion of vehicle;

(1.2) all-links that traversal vehicle node is set up, to each link execution step (2.3), until traversal finishes;

(1.3) transmit queue of recipient's node at traversal link two ends, to each message packets execution step (2.4), until traversal finishes;

(1.4) which kind of region that the vehicle node present position of judgement message receiver is message, processes according to judged result, is specially:

If (1.4.1) target area in message, receives this message, put into ACCEPT queue (message sink queue);

If (1.4.2) Relay Region in message, receives this message, put into CARRY queue (message is carried queue);

If (1.4.3) extraneous areas in message, ignores this message.

(2) cache management sub-process

In the present invention, cache management sub-process may be defined as:

Wherein, ROADCHANGE represents the set of link change, and MANAGEMENT represents the set of cache management activity, comprises adjustment (Change), abandons (Drop) and ignores (None).Concrete steps are:

(2.1) resource that initialization cache management needs, as, obtain message queue, obtain the essential information of state of motion of vehicle, the change type of judgement direction of vehicle movement etc.

(2.2) buffer queue of traversal vehicle node, to each message execution step (2.3), until traversal finishes.

(2.3) if the change type of direction of vehicle movement is oppositely tune, carry out (2.3.1); If the change type of direction of vehicle movement is left and right, turn, carry out (2.3.2); If the change type of direction of vehicle movement still, for keeping straight on, is left intact.

(2.3.1) the residing ACCEPT queue of this message or CARRY queue are exchanged.

(2.3.2) reduce the quota of this message covering path quantity, the message digit that replacement movement tendency calculates is set to the position when vehicle in front, then judges the current region of living in of vehicle node is for which kind of region of this message: if ACCEPT queue is inserted in target area; If Relay Region, inserts CARRY queue; If extraneous areas, deletes this message.

3, last, the present invention can effectively process the accurate speed delivery problem of message under urban road environment, specifically has:

(1) trend type message propagation example

As shown in Figure 3.Take traffic accident message as example, and the target vehicle of message is all vehicles that are about to sail into this position, the target vehicle that all vehicles that trend towards message source position are this message.The target area that the residing region of these vehicles is message.The present invention, according to urban road feature, according to the feature of wagon flow, accurately chooses target area message is spread.Meanwhile, the reverse road in target area arranges Relay Region and is responsible for carrying and preserving of message, contributes to message faster more accurately to cover target area.

(2) deviate from type message propagation example

As shown in Figure 4.Take emergency vehicle information as example, and the target vehicle of message is the vehicle of road the place ahead road occupying, all target vehicles that to deviate from the vehicle of message source position be this message.The target area that the residing region of these vehicles is message.The present invention, equally according to urban road feature, according to the feature of wagon flow, accurately chooses target area message is spread.Meanwhile, the reverse road in target area arranges Relay Region and is responsible for carrying and preserving of message, contributes to message faster more accurately to cover target area.

In addition, can also meet by the setting of configuration messages taxis preference more user's needs, these different configurations are also considered as protection scope of the present invention.

Two,, for making the present invention easier to understand, promote effect more directly perceived, then the present invention is further elaborated in conjunction with a simulation example, but this example does not form any limitation of the invention.

1, simulation configurations

According to the performing step of introducing in embodiment one, emulation platform ONE (referring to Keranen A.Opportunistic network environment simulator[J] .Special Assignment report, Helsinki University of Technology, Department of Communications and Networking, 2008.) on, realize message Multicast Routing Algorithm of the present invention (being called for short CMTG), and under this environment, realize three kinds of relatively agreements: Epidemic Broadcast (being called for short EB) simultaneously, Direct Broadcast (being called for short DB) and DT-IVG are (referring to Bachir A, Benslimane A.A multicast protocol in ad hoc networks inter-vehicle geocast[C] //Vehicular Technology Conference, 2003.VTC2003-Spring.The57th IEEE Semiannual.IEEE, 2003, 4:2456-2460.).The relevant parameter information of simulating scenes is as shown in table 1:

Table 1:ONE simulation configurations parameter

Simulating area size	3km×2km
		Emulation duration	1000s
Vehicle number	500
		Car speed	5～15m/s
Communication radius	50m
		Data transmission rate	256kbps
Vehicle node buffer memory	50MB
		Message size	500kB～1MB

2, simulation result

As shown in Figure 5, in transmission of messages success rate, increase along with simulation time, the transmission success rate of institute's protocols having is all increasing, but in speed, there is Epidemic Broadcast (EB) to be greater than routing algorithm of the present invention (CMTG), be greater than DT-IVG, be greater than Direct Broadcast (DB).And only have CMTG and EB agreement can reach 100% transmission success rate, and DT-IVG and DB agreement all cannot the coverings completely of message within the scope of feasible region in the simulation time of 1000 seconds.

As shown in Figure 6, upper in message overhead rate (message number of (message number of network repeating-reality is delivered successful message number)/network repeating), along with the increase of number of nodes, the overhead rate of institute's protocols having changes little.Wherein, DB agreement is because be single copy agreement, so message overhead rate remains 0; CMTG agreement is because taked copy controlling mechanism, and message overhead rate is little compared with EB and DT-IVG agreement.

As shown in Figure 7, on message dilivery time delay distributes, along with the increase of message dilivery time delay, the sum that the delivery data bag that is less than this time delay accounts for entire packet continues to increase.Visible, the relation of low delay packet proportion has: CMTG agreement is greater than DB, is greater than DT-IVG and EB.

Comprehensive above Performance Ratio, routing algorithm of the present invention, because considered the situation of real road, so be more adapted at organizing in city vehicle network the diffusion of message, it can greatly reduce the overhead of network under the prerequisite of not obvious loss delivery success rate, and can ensure that the message of large number (approximately 80%) is delivered to destination node location within the shorter time (300 seconds) absolutely.

The above is embodiments of the present invention; certainly can not limit with this interest field of the present invention; should be understood that; for those skilled in the art, under the premise without departing from the principles of the invention, can also make some improvement and change; as change the setting of message taxis preference; according to the situation of actual vehicle, set vehicle node cache size, realize etc. on different emulation or actual development platform, these improvement and change are also considered as protection scope of the present invention.

Claims (5)

1. the city vehicle internet message multi-broadcast routing method based on node movement tendency, is characterized in that: specific implementation step is as follows:

(1) whether vehicle node cycle criterion self sets up state or direction of motion variable condition in link, if set up state in link, and execution step (2), if in direction of motion variable condition, execution step (3);

(2) vehicle node and some nodes are set up communication linkage, carry out message sub-process, and concrete grammar is:

(2.1) resource that initial message transmission needs, resource comprises obtains the queue of message sending and receiving, and obtains the essential information of state of motion of vehicle;

(2.2) all-links that traversal vehicle node is set up, to each link execution step (2.3), until traversal finishes;

(2.3) transmit queue of recipient's node at traversal link two ends, to each message packets execution step (2.4), until traversal finishes;

(2.4) which kind of region that the vehicle node present position of judgement message receiver is message, processes according to judged result, is specially:

If (2.4.1) target area in message, receives this message, put into message sink queue, i.e. ACCEPT queue;

If (2.4.2) Relay Region in message, receives this message, put into message and carry queue, be i.e. CARRY queue;

If (2.4.3) extraneous areas in message, ignores this message;

(3) vehicle node seesaws direction while changing, XM cache management sub-process, and concrete grammar is:

(3.1) resource that initialization cache management needs, resource comprises obtains message queue, and obtains the essential information of state of motion of vehicle, the change type of judgement direction of vehicle movement;

(3.2) buffer queue of traversal vehicle node, to each message execution step (3.3), until traversal finishes;

(3.3) if the change type of direction of vehicle movement is oppositely tune, carry out (3.3.1); If the change type of direction of vehicle movement is left and right, turn, carry out (3.3.2); If the change type of direction of vehicle movement still, for keeping straight on, is left intact;

(3.3.1) the residing ACCEPT queue of this message or CARRY queue are exchanged;

(3.3.2) reduce the quota of this message covering path quantity, the message digit that replacement movement tendency calculates is set to the position when vehicle in front, then judges the current region of living in of vehicle node is for which kind of region of this message: if ACCEPT queue is inserted in target area; If Relay Region, inserts CARRY queue; If extraneous areas, deletes this message.

2. the city vehicle internet message multi-broadcast routing method based on node movement tendency according to claim 1, it is characterized in that: which kind of region that the vehicle node present position of the judgement message receiver in described step (2.4) is message, concrete grammar is:

(1) determine that vehicle node position is to the orientation vector v of message source node position ₁, determine vehicle node direction vector v now ₂, obtain the angle theta between two vectors;

(2) according to the directivity feature between two vectors of size judgement of angle theta, be specially:

(2.1) if θ is ∈ [0, π/6], two vectors are considered as in the same way; If θ is ∈ (π/6, π/3), execution step (2.2); If θ is ∈ [π/3,2 π/3], two vectors are considered as intersecting; If θ is ∈ (2 π/3,5 π/6), execution step (2.3); If θ is ∈ [5 π/6, π], two vectors are considered as oppositely;

(2.2) utilize angle between two vectors of Probit model decision-making in the same way or intersect;

(2.3) utilizing the angle between two vectors of Probit model decision-making is intersection or reverse;

(3) according to two vector v ₁and v ₂between directivity feature, obtain the taxis feature tendency of vehicle node and message, be specially:

(3.1) if v ₁with v ₂in the same way, vehicle node trends towards message source node position;

(3.2) if v ₁with v ₂oppositely, vehicle node deviates from message source node position;

(3.3) if v ₁with v ₂intersect, vehicle node has nothing to do in message source node position;

(4) obtain the set TENDENCY of the taxis preference of message _m, judge whether the taxis tendency of vehicle node and message source node position belongs to this set, thereby which kind of region that the vehicle node present position of judgement message receiver is message is specially:

(4.1) if tendency ∈ is TENDENCY _m, the target area in message;

(4.2) if～tendency ∈ TENDENCY _m, the Relay Region in message;

(4.3) otherwise, the extraneous areas in message.

3. the city vehicle internet message multi-broadcast routing method based on node movement tendency according to claim 1, is characterized in that: in described step (3.1), the method for the change type of judgement direction of vehicle movement is:

(1) determine the direction of motion vector v of vehicle node previous moment ₁, determine vehicle node direction of motion vector v now ₂, obtain the angle theta between two vectors;

(2) according to the change type of the size judgement direction of vehicle movement of angle theta, be specially:

(2.1), if θ is ∈ [0, π/6], by the change type of looking direction of vehicle movement, be craspedodrome; If θ is ∈ (π/6, π/3), execution step (2.2); If θ is ∈ [π/3,2 π/3], by the change type depending on direction of vehicle movement, be to turn in left and right; If θ is ∈ (2 π/3,5 π/6), execution step (2.3); If θ is ∈ [5 π/6, π], by the change type of looking direction of vehicle movement, be oppositely tune;

(2.2) utilize the change type of Probit model decision-making direction of vehicle movement for keeping straight on or left and right turning;

(2.3) utilize the change type of Probit model decision-making direction of vehicle movement for turning or oppositely tune left and right.

4. the city vehicle internet message multi-broadcast routing method based on node movement tendency according to claim 2, it is characterized in that: in described step (2.2) and (2.3), utilized Probit model to carry out decision-making, judge the directivity feature between two vectors, Probit model is:

$P_i=\mathrm{Prob}(Y_i=1|x)=\mathrm{Prob}[u_i>-({\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}})]=1-F[{\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}}]$

Wherein, X _ijj independent variable while representing the i time prediction, Y _idependent variable while representing the i time prediction, P _ithe probability that during the i time prediction, dependent variable value is 1, β ₀, β _j, μ _ifor model parameter;

The explanatory variable of choosing is two angle theta between vector, i.e. ANGLE supposes that regression model is:

Y _i＝F ^-1(P _i)＝β ₀+β ₁ANGLE _i+μ _i

Wherein,

Finally, utilize the method for regression analysis, by some actual acquisition to data calculate corresponding parameter, then carry out the corresponding result of decision-making with this model.

5. the city vehicle internet message multi-broadcast routing method based on node movement tendency according to claim 3, it is characterized in that: in described step (2.2) and (2.3), utilized Probit model to carry out decision-making, the change type of judgement direction of vehicle movement, Probit model is:

$P_i=\mathrm{Prob}(Y_i=1|x)=\mathrm{Prob}[u_i>-({\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}})]=1-F[{\mathrm{\β}}_0+\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\mathrm{\β}}_j{X}_{\mathrm{ij}}]$

Wherein, X _ijj independent variable while representing the i time prediction, Y _idependent variable while representing the i time prediction, P _ithe probability that during the i time prediction, dependent variable value is 1, β ₀, β _j, μ _ifor model parameter;

The explanatory variable of choosing is two angle theta between the direction of motion, i.e. ANGLE supposes that regression model is:

Y _i＝F ^-1(P _i)＝β ₀+β ₁ANGLE _i+μ _i

Wherein,

Finally, utilize the method for regression analysis, by some actual acquisition to data calculate corresponding parameter, then carry out the corresponding result of decision-making with this model.